The paper presents the first results of the ongoing INTER-Health pilot coming out from the application of the strategy proposed and developed within the framework of the INTER-IoT (Horizon 2020) European project. The novel IoTbased healthcare platform, supporting decentralized and mobile monitoring of assisted livings in highly heterogeneous contexts, has been obtained by integrating other existing heterogeneous, non-interoperable IoT platforms following the INTER-IoT approach to provide value added assisted living mobile healthcare services (respect to the standard “manual” monitoring performed by conventional healthcare centers) in terms of faster detection and correction of wrong lifestyles or high risk critical situations.
https://edottechnologies.in/All the interoperability aspects and technological choices have been presented to discuss the ease of such integration process, also making easier and faster the interaction between doctor and patient drastically reducing the dropout rate of people leaving the pilot.
One of the most attractive application environment in the IoT arena is healthcare [1]. IoT technologies have the potential to boost up many medical applications, such as remote and mobile health monitoring, human activity recognition, reduction of hospitalizations of chronic patients and elderly care at home. The development of new applications and services can help to decrease healthcare costs, improve the quality of life and enhance the patients’ experience; however, specific healthcare requirements were often faced developing particular and custom systems (i.e. proprietary, closed) that were not interoperable and/or reusable in different, though similar, contexts. Today there is a plethora of IoT systems that are not enough scalable neither sustainable because they are not able to interoperate and whose integration process is almost expensive as the creation of a new solution from scratch. EDOT Technologies As an example, the work in [2], presented an IoT solution, developed from scratch, for monitoring groups of sedentary people by collecting users’ physiological data through wearable devices and sending such data to a computational cloud to be processed, stored and presented to the healthcare professional. This example is a typical custom solution that did not have any benefit from already existing systems and experiences.
Looking at the consumer electronic context, the major manufacturers (e.g. Samsung, Polar, FitBit, Garmin) offer a huge set of devices and applications for human activity recognition paving the way for the emergence of many new service platform (e.g. Runtastic, Strava, Google Fit). However, all these worlds are closed; they do not communicate with each other and create a barrier to the development of new services and platforms. This trend has limited the development of many other IoTbased health care solutions that may constitute a higher addedvalue if we were able to create new services by combining existing ones, benefiting from the interoperability that can become the driving force behind a new development impulse in highly heterogeneous contexts. In fact, many different IoT platforms already exist [3],[4] providing specific solutions for different application scenarios such as healthcare, logistics, home automation, etc... however, the quick integration and interoperability of those heterogeneous IoT platforms, allowing heterogeneous elements to cooperate seamlessly to share data, infrastructures and services as in a homogeneous scenario, is still a challenging issue[5]. It is worth noting that, when we consider any IoT platform, the complexity of the technologies used to build up the platform further arises as each defined layer (device, networking, middleware, application service, data and semantics) exploits specific heterogeneous IoT technological solutions that need to be holistically adapted to form the final platform without providing any (or even limited) interoperability. Thus, it is critical to provide bottom-up “voluntary” approaches able to integrate, interconnect, merge, heterogeneous IoT platforms to build up extreme-scale interoperable ecosystems on top of which large-scale applications can be designed, implemented, executed and managed. Thanks to a novel interoperable approach, proposed and developed within the INTER-IoT [6] H2020 European framework, new cross-platform services can be implemented; hence, to demonstrate the main benefit of the proposed model of integration and interoperability, we created INTER-Health [7], a more powerful IoT healthcare platform for lifestyle monitoring of patients at risk of obesity. This new platform has not been developed from scratch; on the contrary it come out through the interoperability of two existing platforms, UniversAAL [8] and BodyCloud [9] to support new application services that individual platforms were not able to support. The work presented in this paper has been conducted in continuity with respect to the previous one in [10], where we presented the preliminary interoperability approach to support the integration of the two aforementioned IoT healthcare platforms. In this direction, the paper presents the definitive results of the INTER-IoT approach validated through the real implementation of the INTER-Health pilot. All the interoperability aspects and technological choices will be presented to discuss the ease of such integration process also making easier and faster the interaction between doctor and patient. Preliminary medical results will be discussed to demonstrate in which way this new platform can drastically reduce the dropout rate of people leaving the pilot. The rest of the paper is organized as follows. In section II we describe the needs for an integrated IoT-based healthcare platform describing the main components to be integrated and their potentials. After that, we describe the proposed integration approach and the implemented interoperability at middleware layer in section III. The results, on the first six months experimentation of a real INTER-Health pilot involving 200 patients, are presented in section IV to show the improvements in using such integrated platform respect to the current “manual” monitoring performed by conventional healthcare centers. Finally, section V concludes the paper summarizing meaningful considerations and presenting future research directions.
A. Pilot deployment The INTER-Health pilot allows the conduction of an experimental nutritional consulting to reduce health risk factors, at the basis of major chronic diseases, such as incorrect lifestyles which are typically characterized by both high-caloric diet and the lack of an adequate physical activity. In such a way, the pilot involves a group of about 200 patients who want a medical support to improve their lifestyle. Patients can be constantly monitored through a set of wireless wearable devices such as bangles, pedometers and heart rate monitors communicating through standard communication technologies (i.e. Bluetooth for medical devices and wearable sensor nodes); moreover, they have been equipped with smartphones, acting as interoperable gateways [17], toward a cloud repository to be accessed by experts in the specific field, such as doctors, nutritionists and personal trainers, in order to constantly monitor and change the user behaviours. The following objective and subjective measures have been detected through the use of such specific devices: • weight with weekly frequency; • blood pressure, only for those patients that, at the first nutritional counseling, had normal-high pressure values [systolic pressure ≥ 130 and / or diastolic pressure ≥ 85] with daily frequency, morning and evening; • physical activity (number of steps and duration of physical activity practice) with daily frequency; • eating habits and the practice of physical activity twice a week (by online questionnaire). The study, non-invasive and without risk to health has been conducted by the Nutritional Unit (NU) of the Department of Prevention of the Complex Structure of Food, Hygiene and Nutrition of the fifth Local Health Unit of the Italian department of Turin (ASL TO5). During the experimentation, the effectiveness, in terms of lifestyle improvement indexes, of the novel system has been evaluated with respect to the current “manual” monitoring performed by conventional healthcare centers. The presented observational study involved subjects divided into two groups: a group called the Control Group (CG), represented by the patients involved only in the traditional nutritional consulting, and an Experimental Group (EG), represented by the patients that will be involved in the experimental nutritional counseling. In particular, the CG consists of 100 patients (80 females and 20 males) presenting an average age of 47 years while the EG consists of the same number of patients with a similar sex distribution (67 females and 33 males) and average age (46 years). During nutritional counseling, the NU collected and monitored a set of patients data: anthropometric (i.e. weight, height, body mass index, blood pressure, waist circumference), clinical and dietary history (i.e. eating habits and physical activity) for both control and experimental groups. All the collected data have been stored in the electronic health records (EHRs) of the patients and after the first clinical visit, the health status has been constantly monitored through the INTER-Health technological platform. In particular, the nutritional counseling has been performed every three months for the CG and every six months for the EG.
B. Key Performance Indicators: Project Center in Tirunelveli description and measurement In this section, the key performance indicators (KPIs), specifically designed to assess the success of the pilot deployment, are described to show in which way the efficiency and productivity of the overall system is going to be improved through the use of the INTER-IoT components. In particular, we focused our analysis on the following main indicators: • Average BMI improvement • Average waist circumference improvement • Chronic diseases risk reduction • Physical activity (minutes of activity) improvement • Average eating habit improvement • Dropout rate (patients who leave the experimentation) Fig. 7 shows the obtained results after the first six months of experimentation testing the integrated platforms on the EG and making the comparison with the CG without any INTER-Health support. In particular, it is possible to observe that, the apparent increase in the percentage of overweight patients is due to the notable decrease of obese patients which losing weight become overweight. Although both CG and EG experienced a comparable decrease in weight and high risk chronic diseases, the dropout rate is drastically decreased from the 45% of the CG to only 5% in EG testifying the interest of the patients in being constantly monitored through the new technological integrated platform that also assists them and motivates them in the periodic completion of the nutritional and habits questionnaires. All subjects started to practice a correct physical activity because they have been constantly motivated through daily advises suggested by the software application installed on their mobile smartphones. Regarding the eating habit improvement, we verified through the filled electronic questionnaires that all subjects consume 3 main meals a day and 85% of subjects consume 5 portions of fruit and vegetables. Other interesting KPIs have been designed to measure the time spent by the users in using the developed software tools; in particular we verified that the average daily usage of the BodyCloud mobile application is around 10 minutes and the average time in using the PWT application from the doctors is about 2 hours a day .
This work presented the first results of the novel INTERHealth use case through which it is possible to provide, with reduced effort, new assisted living healthcare services thanks to the interoperability between different and already developed IoT-based healthcare platforms. The IoT platforms integration has been implemented and tested by following the INTER-IoT approach and all the new components have been described to highlight the ease of the integration process. The obtained results on real group of patients confirmed how the INTER-Health framework overcomes the traditional methods in the relationship between doctor and patient making easier the interaction, increasing the number of patients that can be assisted and drastically reducing the dropout rate of patients leaving the pilot. Finally, it provides more responsive and effective nutritional advice services extending preventive action, in real time, to a larger population.
